Approaching chemical accuracy using full configuration-interaction quantum Monte Carlo: a study of ionization...

Area of Science:

• Quantum chemistry
• Computational physics
• Materials science

Background:

• Accurate calculation of electronic structures is crucial for understanding chemical properties.
• Full configuration-interaction (FCI) methods provide exact solutions but are computationally intractable for large systems.
• Quantum Monte Carlo (QMC) offers a potential pathway to overcome these limitations.

Purpose of the Study:

• To introduce and validate a new quantum Monte Carlo (QMC) method for calculating exact, full configuration-interaction (FCI) energies.
• To assess the accuracy of this QMC method for elemental systems (Li to Mg) using standard basis sets.
• To evaluate the method's performance in predicting ionization potentials.

Main Methods:

• A novel QMC method employing walker annihilation processes to stochastically sample Slater determinants.
• Calculation of FCI energies for neutral and cationic elements from Li to Mg.
• Systematic investigation using the aug-cc-pVXZ Dunning basis sets up to X=5.

Main Results:

• The QMC method successfully accessed extremely large determinant spaces (over 10^15).
• Calculated ionization potentials were consistently accurate within chemical accuracy, with a minor exception for Sodium (Na).
• Extrapolation schemes showed potential for further improvement but were not uniformly superior to direct large basis set calculations.

Conclusions:

• The developed QMC method provides a viable route to exact FCI energies for realistic systems.
• The method demonstrates high accuracy for ionization potentials, highlighting its potential in quantum chemistry.
• Further basis set improvements for specific elements like Na may enhance predictive power.